We propose a new real-time depth upsampling method based on convolutional neural networks (CNNs) that uses the local context provided by semantic information. Two solutions based on convolutional networks are introduced, modeled according to the level of sparsity given by the depth sensor. While first CNN upsamples data from a partial-dense input, the second one uses dilated convolutions as means to cope with sparse inputs from cost-effective depth sensors. Experiments over data extracted from Kitti dataset highlight the performance of our methods while running in real-time (11 ms for the first case and 17 ms for the second) on a regular GPU.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Real-time Semantic Segmentation-based Depth Upsampling using Deep Learning


    Contributors:


    Publication date :

    2018-06-01


    Size :

    714991 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Image-guided ToF depth upsampling: a survey

    Eichhardt, I. n. | British Library Online Contents | 2017


    NDNet: Narrow While Deep Network for Real-Time Semantic Segmentation

    Yang, Zhengeng / Yu, Hongshan / Fu, Qiang et al. | IEEE | 2021


    Real-Time Semantic Segmentation-Based Stereo Reconstruction

    Miclea, Vlad-Cristian / Nedevschi, Sergiu | IEEE | 2020

    Free access

    Guided depth upsampling for precise mapping of urban environments

    Wirges, Sascha / Roxin, Bjorn / Rehder, Eike et al. | IEEE | 2017


    Semantic image segmentation network based on deep learning

    Chen, Bo / Zhang, Jiahao / Zhou, Jianbang et al. | SPIE | 2020